--- title: I Made a Chrome Extension! author: Tris date: 2026-05-16 url: https://www.trishussey.com/i-made-a-chrome-extension/ --- And You Can Too. For free. And it's not hard at all. Towards the end of last year, I had this crazy idea to automate finding and storing articles I think are interesting into a NotebookLM notebook. Well, that plan didn’t actually work out. What *did* work out, though, was making my first Chrome extension. It only took about 45 minutes to make the first version. I thought it was pretty good, good enough to go into the Web Store, but I kept putting it off. I’m glad I did. I took a hard look at it, asked Antigravity to refactor the code, design a better UX-UI, and polish off the rough edges. So now I’m excited that today that my Chrome extension—[LinkCollector](https://chromewebstore.google.com/detail/iaelenjbickdahcnneoccilkkcjimckg?utm_source=item-share-cb)—is live in the Chrome Web Store. [](https://chromewebstore.google.com/detail/iaelenjbickdahcnneoccilkkcjimckg?utm_source=item-share-cb) So, let me tell you a little bit about how I got there, why I built it, and what it means for all of us—especially marketers. And I’ll tell you a little bit about the Chrome extension itself (because I think it’s pretty cool) **I didn’t start out to make a link collecting extension** Like I said at the beginning, I had this idea to try to organize my info library a little better using NotebookLM. I find lots of articles—lots of tabs open at once—things I want to read or save for later. I thought, “I wish, I could store all these in NotebookLM, but how could I make it an easy one click action?” I thought I could easily code something to indicate a tab, take that URL, and send it over to a specific NotebookLM notebook. It would then do its thing in that notebook, which would become a stash and store of interesting articles and information. Not so much. NotebookLM doesn’t have an official API. There is a solid tool for connecting NotebookLM to an MCP, etc., but last December I wasn’t ready to dive into something like that just yet. What ended up happening was I made a really simple Chrome extension. Its sole goal was: you’re on a page, you click a button, and it saves the link to that page. When you’re ready, you can copy all the links to paste into the sources box in NotebookLM. Simple. Easy. A one trick pony with a really good trick. A few people tried it in its early dev state, but I kind of just let it be for a long time. Then I was looking at it and thought, “If I really want to put this into the Chrome web store, it needs to be better.” It needs a better UI and a few more features. I wanted to add file export—Markdown and CSV, hotkeys, options for saving things. Just make it a more polished thing that people would enjoy using it. Because I vibe coded it and I can’t code my way out of a wet paper bag, I had no idea how good or bad the code was, so I had a code review, and it seems to work. It’s a lot peppier than it was before, and it looks a lot better. I’m actually quite happy with it. But that’s not the point here. Here’s the lesson: *If I can do this, you can do this*. I’m going to be publishing a companion to this post about getting started with vibe coding, [building on an earlier post on getting started with Antigravity](https://30plusdaysofai.substack.com/p/day-27-vibe-coding-with-antigravity), including all the things I’ve learned in the three months since I wrote that post back in February. **I did it all with free accounts** I built most of this whole Chrome extension using free accounts. I used my Gemini Pro account to make the original product requirements document (PRD), but I could have done it with a free Gemini account in Pro mode. It might have taken up all my usage for the day, but it certainly could have been done. All of the coding in Antigravity was, until very recently when I started paying for AI tools in my personal Gmail account, was done with free accounts.. Yes, I would run out of usage. And yes, I had to switch between three different Google IDs in Antigravity, logging out and logging back in to tap into the usage of each of those, but I did it. My code is managed on a free GitHub account—I’m far away from needing a paid account I think. The entire point of this example is, if you have an idea for something, you can build it, even if it’s just for you. Even if it’s not awesome by commercial software standards, if it works and works for you, that’s great. If it happens to be useful for other people and you can expand and improve it, super. Beyond that, the rest doesn’t matter. If it works for you, the job is done. It’s not hard. You can stop and start as you have time—AIs don’t care if you leave them waiting for you to answer a question. The first step is just sitting down and talking it out with your AI of choice to make that first PRD and go from there (I’ll build on this in my post soon). **This is what’s next, and it’s awesome** We are at this amazing inflection point, very much like when people started being able to put up their own websites, where if you have an idea, you can build it. I know people are talking about getting billion-dollar startups with a single person coding the whole thing with a small amount of money spent on APIs, and cloud storage, but this is more practical: you have an idea; you need a tool to do a thing—just build it. Want to build a prototype of a landing page or a site or whatever? Want a simple little script to clean up data or organize a folder? No problem. Pretty much any task you can use a computer for now, you can code up in short order and have running. You can do any of it, and you can do it for free. That’s the story. I wanted a better way to collect links to send to NotebookLM, and I built it. [](https://chromewebstore.google.com/detail/iaelenjbickdahcnneoccilkkcjimckg?utm_source=item-share-cb) [Link Collector is available now on the Chrome web store](https://chromewebstore.google.com/detail/iaelenjbickdahcnneoccilkkcjimckg?utm_source=item-share-cb). I hope you give it a shot. People have already tried it, and I’ve got my first couple of feature requests—which is really exciting—so those are going in the queue. I’m going to take these ideas to map out the next version and send it for approval. In the meantime, ask yourself: What do I want to build? Then build it. And watch for my guide to coding and vibe coding coming up soon.